Avoiding Brittleness with Metacognitive Loops

Posted 11 Feb 2005 at 18:06 UTC by steve Share This

Unlike a human, even the best designed robot may fail when it faces a problem it wasn't specifically designed to solve. This is known as brittleness and AI researchers consider it one of the most challenging problems of AI. Michael L. Anderson and Donald R. Perlis have released a new paper, titled "Self-awareness and self-improvement: The metacognitive loop and the problem of brittleness" (PDF format), that poses a solution to the problem. The metacognitive loop is basically a process by which a machine can "notice something is amiss, assess the anomaly, and guide a solution into place".

See more of the latest robot news!

Recent blogs

28 Nov 2015 mwaibel (Master)
25 Nov 2015 shimniok (Journeyer)
6 Nov 2015 wedesoft (Master)
26 Oct 2015 steve (Master)
20 Oct 2015 Flanneltron (Journeyer)
10 Sep 2015 svo (Master)
9 Aug 2015 Petar.Kormushev (Master)
6 May 2015 spirit (Journeyer)
14 Nov 2014 Sergey Popov (Apprentice)
3 Jul 2014 jmhenry (Journeyer)
Share this page